PROSPECT THEORY The objective of prospect theory is to

describe how people make decisions when

ANDREW CHIU there is uncertainty about the consequences
GEORGE WU of their choices. Decision theorists distin-
University of Chicago Booth guish between decision under risk, situations
School of Business, in which the likelihood of events are known
Center for Decision or objective (such as a spin of a roulette
Research, Chicago, Illinois wheel), and decision under uncertainty,
situations in which the decision maker
must assess the probability of the uncertain
events and hence the likelihood of events are
Prospect theory is a descriptive theory of how
subjective (such as the outcome of a sports
people make choices that involve risk. The
game) [3]. Although prospect theory applies
theory was developed by psychologists Kah-
to both risk and uncertainty, we will focus
neman and Tversky in 1979 [1]. In 2002,
here on risk for simplicity.
Kahneman was awarded the Nobel Prize
in Economics, partially due to prospect the-
ory’s enormous influence on economics and
other social sciences, and partially due to EXPECTED UTILITY THEORY
Kahneman and Tversky’s research on heuris-
tics and biases [2]. (The Nobel Prize can- Before prospect theory, the dominant theory
not be awarded posthumously, and Tversky of how people choose among risky alterna-
died in 1996.) Prospect theory challenged tives was expected utility theory. This theory
the traditional view in economics that deci- was originally proposed by Bernoulli in 1738
sion makers are rational and, in particu- [4] and was formalized by von Neumann and
lar, expected utility maximizers. However, Morgenstern in 1947 [5]. von Neumann and
prospect theory’s reach has not been limited Morgenstern proposed a set of axioms, or
to economics—the theory has been influ- basic conditions, that are necessary and suf-
ential in understanding a wide variety of ficient for expected utility. Expected utility
real-world phenomena, in fields ranging from theory is a standard building block of modern
business, law, and medicine to political sci- economic theory [6,7]. It is regarded to be
ence and public policy. rational to be an expected utility maximizer,
Most real-world decisions must be made as this theory is based on compelling axioms
without full knowledge of what will transpire about how people should behave [8].
in the future. Consider, for example, an oil Expected utility theory posits that deci-
and gas company debating whether to drill for sion makers choose the prospect that max-
oil, an investor deciding whether to purchase imizes their expected (or average) utility.
Microsoft stock, or a family choosing whether More formally, consider a risky prospect, P =
to vacation in Chicago in May. These deci- (p1 , x1 ; . . . ; pi , xi ; . . . ; pn , xn ), that offers
 out-
sions would be straightforward if the decision come xi with probability pi , where pi = 1.
makers could predict the future with cer- According to expected utility,  the prospect
tainty. But people are not clairvoyant. There P is evaluated as U(P) = ni=1 pi u(xi ), with
may or may not be oil in the ground. Investing decision makers choosing the prospect with
in Microsoft stock may increase your wealth, the highest expected utility.
but you may also end up losing a sizable por- Under expected utility, risk preferences
tion of your initial investment. Weather in are completely captured by the shape of the
Chicago in May could be glorious or unbear- utility function. Decision makers are risk
able. averse if U(x) is concave, and risk seeking

Wiley Encyclopedia of Operations Research and Management Science, edited by James J. Cochran
Copyright © 2010 John Wiley & Sons, Inc.
1
 10.1002/9780470400531.eorms0687, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/9780470400531.eorms0687, Wiley Online Library on [06/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2 PROSPECT THEORY

ifU(x) is convex, with most classical economic But individuals are not universally risk
theory based on the tenet that decisions averse. They dislike risk in some situations,
makers are risk averse. One standard while liking risk in others. For example,
interpretation for risk aversion is that the lotteries and casinos are billion-dollar indus-
usefulness of an additional dollar decreases tries in the United States. Many individuals
as a person gets wealthier, a principle known purchase lottery tickets, even though the
as diminishing marginal utility [9]. A utility expected value of a state lottery ticket is
function that exhibits diminishing marginal often as little as half the price of the lottery
utility is concave, and hence the decision ticket [10].
maker is risk averse. Kahneman and Tversky demonstrated
that the purchase of lottery tickets is not an
isolated example of risk-seeking behavior.
PROSPECT THEORY
They documented a pattern that has become
known as the fourfold pattern of risk atti-
Kahneman and Tversky showed that deci-
tudes [11]. Individuals are risk averse for
sion makers are not rational as required
by expected utility theory. The theory has most gains, but risk seeking for most losses.
been successful in making this point, largely As previously mentioned, most people would
because the theory consists of three parts choose $500 for sure over a lottery that
that are individually attractive and work offered them an equal chance at $1000 or $0.
together well. First, Kahneman and Tver- On the other hand, the same person typically
sky produced a number of intuitive and ele- prefers an even chance at losing $0 or losing
gant empirical demonstrations that were at $1000 to losing $500 for sure. This pattern,
odds with expected utility theory. Second, however, switches when probabilities are
they proposed a mathematical model that small. In this case, decision makers are
could explain and organize these violations. risk seeking for gains, and risk averse for
Finally, they posited some psychological prin- losses. Consider a choice between winning
ciples that suggest how people evaluate the $1 for sure and a prospect that offered a 1%
alternatives they face. We begin by reviewing chance at winning $100. Gambling in this
the primary empirical demonstrations that case seems pretty reasonable. After all, $1 is
challenged the standard theory that deci- not much money, and the lottery offers some
sion makers are rational. We then describe chance at winning a sizable amount of cash.
the pieces of the mathematical model used The purchase of a lottery ticket is indeed an
to organize these violations. Throughout, we example of risk seeking for small probability
discuss how the psychological principles pro- gains. However, preferences reverse when
vide insight into these empirical demonstra- we change ‘‘winning’’ to ‘‘losing’’ in the above
tions and review these principles at the end example. Now, the typical preference is for
of this section.
losing $1 for sure over taking a prospect in
which there is a 1% chance of losing $100.
Empirical Demonstrations The attractiveness of insurance provides an
Fourfold Pattern of Risk Attitudes. Stan- example of risk aversion for small probability
dard economic theory assumes that individu- losses. In sum, the fourfold pattern of risk
als are risk averse. Consider a lottery that preferences is risk aversion for medium
offers a 50% chance to win $1000 and a to large probability gains, risk seeking for
50% chance to win $0. The expected value of small probability gains, risk seeking for
this lottery is $500 (0.5 × $1000 + 0.5 × $0). medium to large probability losses, and
A risk-averse individual prefers winning the risk aversion for small probability losses.
expected value of this lottery, $500, for cer- This pattern is often called the reflection
tain, to the lottery itself. Indeed, surveys effect, because preferences flip or ‘‘reflect’’
show that the vast majority of people are risk when outcomes are changed from gains to
averse for this choice. losses.
 10.1002/9780470400531.eorms0687, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/9780470400531.eorms0687, Wiley Online Library on [06/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
PROSPECT THEORY 3

Framing. Expected utility theory assumes Aversion to Losses. Expected utility

that preferences between prospects do not theory assumes that choices only reflect
depend on the manner in which they are final outcomes. For example, if one were
described. For example, $20 should be the beneficiary of a $100 check, but also
viewed the same whether you are given a received a $100 speeding ticket, these two
single $20 bill, two $10 bills, or 20 singles. events would offset one another in monetary
This compelling principle is known as the terms. However, Kahneman and Tversky
invariance assumption. However, prospect demonstrated that individuals tend to dislike
theory demonstrates that the same options losses much more than they like gains. Thus,
can be framed in different ways to produce the two events do not offset one another in
dramatically different preferences. In other psychological terms. The $100 check makes
words, our choices do not always obey the you happy, to be sure. But, your pleasurable
invariance assumption. feelings for the check pale in comparison with
To illustrate, consider the following two- the pain you feel when you are reminded
choice situations: about your $100 speeding ticket. This dislike
of losses is known as loss aversion. Put
simply, losses loom larger than gains.
• In situation 1, you are first given $1000. To illustrate loss aversion, consider a
You must now choose between two gamble in which you may lose $1000 or
options. If you choose A, you will receive gain $1100 with equal chance. Although the
an additional $500 for sure. If you expected value of this gamble is positive,
choose B, there is an equal chance that most people find this prospect unappealing.
you will receive either an additional The potential loss of $1000 cannot be offset
$1000 or nothing. by the potential gain of $1100. In order to
• Now consider situation 2; this time you make this gamble attractive, the gain has to
are first given $2000. Again, there are be increased considerably, to about $2000. In
two options. Perhaps you would like C, a other words, roughly speaking, the pain of a
sure loss of $500? Or, maybe you would loss is about twice as much as the pleasure
prefer D, a 50% chance at losing $1000 from a comparably sized gain [12].
and a 50% chance at losing $0?

Common-Consequence Effect. Suppose

When confronted with situation 1, most that you are faced with the following choice.
people prefer A, the sure $500, to the gamble If you choose A, you receive $200 with 10%
B. On the other hand, when posed with a chance. If you choose B, you receive $100 with
choice between C and D, most choose D, the 20% chance. Next, imagine that we improve
uncertain loss, over the sure loss C. While both options by adding to each a 10% chance
both choices seem reasonable in isolation, of receiving $1000. Now if you choose A, you
situations 1 and 2 are identical in terms receive $200 with 10% chance, $1000 with
of final consequences, reducing to a choice 10% chance, and $0 with 80% chance. If
between $1500 for sure (A and C) and a you choose B, you receive $100 with 20%
lottery that offers an even chance at $1000 chance, $1000 with 10% chance, and $0 with
or $2000 (B and D). 70% chance. It seems reasonable that your
The standard preferences described above choices in these two situations be identical,
violate the invariance assumption invoked that is, unaffected by the additional chance
by expected utility. However, they are con- at receiving $1000. Since both of the sec-
sistent with the risk attitudes revealed by ond options have a 10% chance of receiving
the fourfold pattern of risk attitudes. When $1000, this common consequence is irrele-
the problem is framed as a gain (situation 1), vant for this choice. Expected utility the-
people are usually risk averse. But when the ory assumes this principle—adding a com-
problem is framed as a loss (situation 2), mon consequence to two prospects should not
people tend to be risk seeking. change which alternative the decision maker
 10.1002/9780470400531.eorms0687, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/9780470400531.eorms0687, Wiley Online Library on [06/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4 PROSPECT THEORY

prefers. This principle is known as the inde- sure or (B) a 0.10 chance at $5 million, a
pendence axiom [13,14]. 0.89 chance at $1 million, and 0.01 chance
Kahneman and Tversky documented a vio- at $0, whereas the second was a choice
lation of this principle. Consider the following between (C) a 0.11 chance at $1 million (and
two choices: a 0.89 chance at $0) and (D) a 0.10 chance
at $5 million (and a 0.90 chance at $0).
Situation 1 Most people prefer A and D, a pattern that
– option A offers a 33% chance at $2500 contradicts expected utility. Under expected
and a 67% chance at $0; utility theory, a choice of A over B implies
that 0.11u(1) > 0.10u(5) + 0.01u(0), whereas
– option B offers a 34% chance at $2400
and a 66% chance at $0. a preference for D over C implies the opposite
inequality, 0.10u(5) + 0.01u(0) > 0.11u(1).
Situation 2
– option C offers a 33% chance at
$2500, a 66% chance at $2400, and a The Prospect Theory Model
1% chance at $0; Prospect theory offered a rich set of empirical
– option D offers $2400 for sure. challenges to expected utility theory, includ-
ing the fourfold pattern of risk attitudes,
Most people prefer A to B, arguing that 33% framing effects, aversion to losses, and the
and 34% seem about the same, but $2500 common-consequence effect. These demon-
is clearly better than $2400. On the other strations are robust, attractive, and intuitive.
hand, people like D over C, reasoning that it But do these demonstrations inform us about
is not worth taking a chance with C since D how individuals make decisions that involve
has a sure chance of winning a good amount risk in general? Kahneman and Tversky pro-
of money. To see that this pattern of choices posed a model for organizing the empirical
violates the independence axiom and hence is demonstrations described above and making
inconsistent with expected utility, note that more accurate predictions about how individ-
situation 2 is created from situation 1 by uals would behave in other choice situations.
adding a common consequence to both A and Like expected utility theory, prospect
B: a 66% chance at $2400. theory assumes that individuals evalu-
This problem illustrates the strong prefer- ate each alternative and then choose the
ence for certainty over uncertainty. Indeed, alternative with the highest subjective
this example has been termed the certainty valuation. However, prospect theory differs
effect. Adding a 66% chance at $2400 to B, a from expected utility theory in two sub-
34% chance at $2400, is particularly attrac- stantive ways. Suppose a decision maker
tive, because it changes the prospect from might choose an alternative that offers a p
possible to certain. On the other hand, for chance at outcome x. Expected utility theory
option A, the additional 66% chance at $2400 assumes that the utility of this alternative is
increases the chance of winning from 33% to pu(x). Prospect theory, like expected utility,
99%, or from probable to more probable. assumes that the decision maker evaluates
This demonstration is also called the the x that he/she might get and the p chance
common-consequence effect, because adding a at winning, and then combines these two
common consequence to two options changes evaluations together. However, the valuation
preferences, contrary to expected utility of outcomes is governed by the value function,
theory [15]. Indeed, a variant of this problem v(x), and the valuation of the probabilities is
posited by the French economist Maurice governed by the probability weighting func-
Allais was one of the earliest challenges to tion, π (p), and thus the value of this alter-
the descriptive validity of expected utility native is given by π (p)v(x). [The valuation
[16]. Allais’ challenge, which has become of more complicated alternatives with more
known as the Allais Paradox, consisted of than one nonzero outcome follows a revision
two pairs of gambles. The first involved of prospect theory, cumulative prospect theory
a choice between either (A) $1 million for [11].] These two pieces are described below.
 10.1002/9780470400531.eorms0687, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/9780470400531.eorms0687, Wiley Online Library on [06/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
PROSPECT THEORY 5

Value Function. The prospect theory value

function, v(x), captures an individual’s sub-
jective perception of a particular outcome
(Fig. 1). Prospect theory assumes that out-
comes are coded as gains and losses, or out-

v(x)
comes above and below a reference point,
and the value function captures how much
better one gain is than another gain, how
much worse one loss is than another loss, or
alternatively how a particular loss compares
with a gain of the same magnitude.
Three psychological principles constrain
the shape of the value function. The first
principle, reference dependence, suggests that
outcomes are viewed relative to a reference x
point and hence coded as gains or losses.
Figure 1. A typical prospect theory value function,
A $10,000 bonus is seen differently if an
v(x), which is concave for gains and convex for
employee is expecting a $5000 bonus or a losses (and thus consistent with the principle of
$15,000 bonus. The bonus is seen as a $5000 diminishing sensitivity) and is steeper for losses
gain in the first case, but a $5000 loss in the than gains (and thus exhibits loss aversion).
second case.
The second principle, diminishing sen-
sitivity, borrows from research on the preference is also consistent with diminish-
psychophysics of perception. Research has ing sensitivity, because the first $500 loss is
shown that an individual is highly likely to much more painful than the second $500 loss.
discriminate between a 2 and 3 kg weight, Thus, the value function in Fig. 1 is concave
but not very likely to notice the difference for gains and convex for losses.
between 22 and 23 kg weights [17]. Anal- The third psychological principle under-
ogous findings appear in tasks ranging lying the value function is loss aversion.
from the perception of line lengths to the Recall that losses loom larger than gains.
perception of temperature. Prospect theory A loss of $100 seems much more painful
than a gain of $100 seems pleasurable. Most
thus suggests that changes in value have a
people dislike a prospect that gives an equal
greater impact near the reference point than
chance at winning $1000 or losing $1000.
away from the reference point. Thus, there
These two examples are captured by a value
is a big difference between a $100 gain and
function that is steeper for losses than gains
a $200 gain, but a much smaller difference
(−v(−x) > v(x)), as is depicted in the value
between gains of $1100 and $1200. Similarly,
function in Fig. 1.
a loss of $100 seems quite distinct from a
loss of $200, but losses of $1100 and $1200 Probability Weighting Function. Prospect
seem pretty similar. theory assumes that individuals do not
The principle of diminishing sensitivity weigh outcomes by their probability, as
away from the reference point is consistent in expected utility theory, but by some
with the reflection effect described above. distortion of probabilities. This distortion of
Most people would choose $500 for sure over probability is captured by prospect theory’s
an even chance at winning $1000 or winning probability weighting function, π (p) (Fig. 2).
$0. Diminishing sensitivity suggests that the Interestingly, the psychology governing the
first $500 is more pleasurable than the second distortion of probability involves two of the
$500. When this choice is changed to losses, three principles used to understand the value
however, preferences switch. The majority of function, reference dependence and dimin-
individuals prefers an even chance at losing ishing sensitivity away from a reference
$1000 or losing $0 to losing $500 for sure. This point. For probability, there are two obvious
 10.1002/9780470400531.eorms0687, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/9780470400531.eorms0687, Wiley Online Library on [06/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
6 PROSPECT THEORY

1.0 Diminishing sensitivity implies that low

0.9 probabilities are typically given more weight
0.8
than they would receive using expected
utility. This overweighting is consistent
0.7
with risk seeking for low probability gains
0.6 (such as lottery tickets) and risk aversion for
π(p)

0.5 low probability losses (such as insurance).

0.4 Medium to high probabilities are typically
0.3 given less weight than they would receive
using expected value. Such underweighting
0.2
is consistent with risk aversion for medium
0.1
to high probability gains, and risk seeking
0.0 for medium to high probability losses. The
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
probability weighting function also helps us
p
understand the common-consequence effect
Figure 2. A typical prospect theory probability described above. In the first choice, the
weighting function, π (p), which is concave for small probabilities of winning a prize are 33% and
probabilities and convex for medium to large prob- 34%, whereas the probabilities of winning
abilities (and thus consistent with the principle of with the second choice are 99% and 100%.
diminishing sensitivity). The probability weighting function suggests
that the first difference is trivial (thus
reference points, certainty and impossibility, decision makers look at outcomes to make
or 100% chance and 0% chance. The distor- this choice), whereas the second difference
tion of probability shown in the probability is consequential (and thus sufficient to drive
weighting function captures the diminishing the decision).
sensitivity away from these two reference
points. People are most sensitive to changes Psychological Principles
in probability when they are near 0% or 100% Prospect theory invokes three psychologi-
than when the change applies to intermedi- cal principles: reference dependence, dimin-
ate probabilities. Consider adding a chance ishing sensitivity, and loss aversion. The
to win a large prize, such as a new car. first two principles produce the characteristic
Which of the following appears to be the least form of the value function and the probability
significant increase: increasing your chance weighting function shown in Figs 1 and 2.
of winning from 0% to 1%, from 33% to 34%, For the value function, outcomes are viewed
or from 99% to 100%? Most people regard the relative to a reference point, with outcomes
change from 0% to 1% as significant, because above the reference point coded as gains and
it changes the chance of winning from outcomes below the reference point coded
impossible to possible [18]. The change from as losses. The valuation of both gains and
99% to 100%, too, seems noteworthy, because losses exhibits diminishing sensitivity away
it shifts the odds from likely to certain. In from the reference point. For the probability
contrast, the change from 33% to 34% seems weighting function, probabilities are viewed
negligible. Put simply, there is a categorical relative to the reference points of impossi-
difference between impossible and possible or bility (0) and certainty [1]. Changes away
between likely and certain, but the difference from these two reference points also exhibit
between possible and slightly more possible is diminishing sensitivity.
a matter of degree. The probability weighting
function in Fig. 2 exhibits diminishing sen-
sitivity: π (0.01) − π (0) > π (0.34) − π (0.33) APPLICATIONS
and π (1) − π (0.99) > π (0.34) − π (0.33).
The function is thus concave for small Prospect theory has been useful for under-
probabilities and convex for medium and standing real-world phenomena in a variety
large probabilities. of domains. A few applications are discussed
 10.1002/9780470400531.eorms0687, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/9780470400531.eorms0687, Wiley Online Library on [06/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
PROSPECT THEORY 7

below to provide a flavor of the broad reach to $22,300. It is clear that the salesperson
the theory has had. For a description of other will be more effective at convincing you to
applications, see Ref. 19. purchase the upgrade if he/she frames the
upgrade as an increase in price. Again, the
principle at work is diminishing sensitivity
Mental Accounting
of the value function.
Although prospect theory was developed to Consider one final example of mental
explain how people make choices involving accounting. Imagine that you can walk
risk, the principles of prospect theory have 10 minutes to save $5. Would you do so?
been used to understand how people choose People give different answers if the purchase
among certain alternatives as well. One is a $25 calculator or a $125 jacket. A $5
influential application of prospect theory is discount seems more significant when it
known as mental accounting [20,21]. Mental is applied to the inexpensive calculator
accounting is a theory of how individuals and rather than the expensive jacket. Of course,
households make consumer and financial this behavior is not rational, since the
decisions. Most of the implications of mental decision to walk 10 minutes boils down to
accounting follow from the shape of prospect a choice of whether walking 10 minutes is
theory’s value function. worth $5.
Mental accounting suggests that people
prefer to segregate gains and aggregate Status Quo Bias
losses. Consider, for example, two pleasur- In many situations, the decision maker must
able outcomes, a $100 check from your aunt decide whether to choose a particular alter-
and $50 won in a school raffle. Most people native or stick with the status quo. How-
find it more enjoyable to view these outcomes ever, the status quo is frequently chosen
separately, $100 plus $50, than together, much more often than it should be if the
$150. The attractiveness of segregating evaluation of the alternative were made sim-
gains follows from the shape of the value ply based on the positive and negative fea-
function and the psychological principle of tures of that alternative. One study found
diminishing sensitivity away from the refer- a strong status quo bias for the choice of
ence point of zero. Next consider two painful automobile insurance plans. Prior to deregu-
outcomes, a $100 speeding ticket and a $150 lation in the early 1990s, Pennsylvania had
tax liability. It seems clear that aggregating an expensive insurance plan that allowed a
the losses is a good idea psychologically. driver to sue another driver in many sit-
In other words, a loss of $250 seems less uations. New Jersey had a cheaper plan
painful than two separate losses of $100 that permitted its driver to sue only in lim-
and $150. Mental accounting suggests that ited situations. As a result of deregulation,
aggregating losses is more attractive than citizens of Pennsylvania were permitted to
segregating losses. This simple principle trade down to the cheaper plan, and peo-
helps us understand a number of real-world ple in New Jersey were allowed to trade up
phenomena, including the attractiveness of to the more expensive plan. Nevertheless,
flat-rate pricing plans. Most people find it 75–80% of drivers stuck with their original
less painful to pay $70 for a monthly health plan [23].
club membership than $10 each time they How does prospect theory help us under-
work out [22]. stand why people tend to prefer the sta-
Mental accounting also suggests that dif- tus quo? When giving up the status quo
ferent ways of framing the same activity for another alternative there are, typically,
can be viewed very differently. Suppose that trade-offs to be made. An option will have
you have decided to purchase a new car for advantages and disadvantages relative to
$22,000 and are contemplating purchasing a the status quo. The advantages are likely
stereo upgrade. A car salesperson can frame to be perceived as gains relative to the sta-
the stereo as a $300 purchase or an increase tus quo, with the disadvantages perceived
in the total price of the new car from $22,000 as losses. Because of loss aversion, losses
 10.1002/9780470400531.eorms0687, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/9780470400531.eorms0687, Wiley Online Library on [06/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
8 PROSPECT THEORY

appear more significant than gains of com- plan, or a plan that will be chosen if the
parable magnitude. The principle that losses employee does not make a choice. This plan
loom larger than gains explains why peo- is often a sensible plan, but it is often chosen
ple have a tendency to remain at the status by the vast majority of employees, even
quo—a phenomenon known as the status quo when another plan might make more sense
bias. [28]. To see why prospect theory can explain
A related demonstration, called the this choice, we assume that the default
endowment effect, shows once again the plan is the reference point for evaluating
asymmetry between losing and gaining another retirement plan. A second plan will
[24]. In one study, half of the participants be better on some dimensions and worse on
are given a lottery ticket and half of the other dimensions. Because of loss aversion,
participants are given $2 [25]. In each the dimensions on which the second plan
case, the participants can keep what they is inferior loom large compared to the
were given or trade it for what the others dimensions on which it is superior, providing
were given. However, very few participants a substantial advantage to the default
chose to switch. Those who were given plan.
lottery tickets largely stuck with the lottery
tickets, but those who were given the cash
REFERENCES
mostly elected to keep the cash. Once again,
prospect theory’s principle of loss aversion 1. Kahneman D, Tversky A. Prospect theory: an
can explain this phenomenon. When people analysis of decision under risk. Econometrica
are endowed with an item, it becomes the 1979;47(2):263–291.
reference point from which other options 2. Rabin M. The Nobel Memorial Prize for Daniel
are evaluated. Since losses loom larger than Kahneman. Scandanavian Journal of Eco-
gains, giving up an item that one possesses nomics 2003;105(2):157–180.
is perceived as more painful than acquiring 3. Knight FH. Risk, uncertainty and profit. New
an item of comparable magnitude that one York: Houghton-Mifflin; 1921.
does not already possess. 4. Bernoulli D. Specimen theoriae novae de men-
sura sortis. Comment Acad Sci Imp Petrop
1738;5:175–192.
Other Applications 5. von Neumann J, Morgenstern O. Theory of
Prospect theory provides insights into games and economic behavior. 2nd ed. Prince-
ton (NJ): Princeton University Press; 1947.
numerous other real-world decisions. We
sketch a few applications below to give a 6. Varian HR. Microeconomic analysis. New
York: W.W. Norton; 1992.
flavor of the power of prospect theory. First,
in horse track betting, favorites are usually 7. Pratt JW. Risk aversion in the small and in the
large. Econometrica 1964;32(1/2):122–136.
underbet, and longshots are typically overbet
[26]. This behavior can be explained by 8. Raiffa H. Decision analysis: introductory lec-
tures on choice under uncertainty. New York:
the probability weighting function. Second,
Addison-Wesley; 1968.
consider an investor who bought two stocks
9. Stigler GJ. The development of utility theory
at $50. One has increased in value to $75,
II. J Polit Econ 1950;58(5):373–396.
and the other has fallen in price to $25.
10. Clotfelter CT, Cook PJ. Selling hope: state lot-
If the investor has to sell one stock, which
teries in America. Cambridge (MA): Harvard
would he/she sell? Financial economists University Press; 1989.
have shown that investors overwhelmingly
11. Tversky A, Kahneman D. Advances in
sell the winner rather than the loser [27]. prospect theory: cumulative representation
The disposition effect suggests that sell- of uncertainty. J Risk Uncertain 1992;5(4):
ing the loser is unattractive, because the 297–323.
investor must come to terms with his/her 12. Abdellaoui M, Bleichrodt H, Paraschiv C.
losses. Loss aversion under prospect theory: a
Finally, we consider choices of retirement parameter-free measurement. Manage Sci
plans. Employers often provide a default 2007;53(10):1659–1674.
 10.1002/9780470400531.eorms0687, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/9780470400531.eorms0687, Wiley Online Library on [06/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
PROSPECT THEORY 9

13. Machina MJ. Choice under uncertainty: prob- 24. Kahneman D, Knetsch JL, Thaler RH.
lems solved and unsolved. J Econ Perspect Experimental tests of the endowment effect
1987;1(1):121–154. and the Coase Theorem. J Polit Econ
14. Samuelson P. Probability, utility, and 1990;98:1325–1348.
the independence axiom. Econometrica 25. Knetsch JL. The endowment effect and evi-
1952;20:670–678. dence of nonreversible indifference curves. Am
15. Wu G, Gonzalez R. Curvature of the Econ Rev 1989;79(5):1277–1284.
probability weighting function. Manage Sci 26. Thaler RH, Ziemba W. Anomalies: parimutuel
1996;42(12):1676–1690. betting markets: racetracks and lotteries. J
16. Allais M. Le comportement de l’homme Econ Perspect 1988;2(2):161–174.
rationel devant le risque, critique des postu- 27. Odean T. Are investors reluctant to realize
lates et axiomes de l’ecole Americaine. Econo- their losses? J Finance 1998;53(5):1775–1798.
metrica 1953;21(4):503–546. 28. Madrian BC, Shea DF. The power of sug-
17. Stevens SS. On the psychophysical law. Psy- gestion: inertia in 401(k) participation and
chol Rev 1957;64:153–181. savings behavior. Q J Econ 2001;116(4):
18. Gonzalez R, Wu G. On the shape of the 1149–1187.
probability weighting function. Cognit Psychol
1999;38(1):129–166.
19. Camerer CF. Prospect theory in the wild: evi-
FURTHER READING
dence from the field In: Kahneman Daniel, Trepel C, Fox CR, Poldrack RA. Prospect the-
Tversky Amos, editors. Choices, values and ory on the brain? Toward a cognitive neuro-
frames. New York: Cambridge University science of decision under risk. Cogn Brain Res
Press; 2000. pp. 288–300. 2005;23(1):34–50.
20. Thaler RH. Mental accounting and consumer Wakker PP. Decision under risk Prospect theory
choice. Mark Sci 1985;4(3):199–214. for risk and ambiguity. Cambridge: Cambridge
21. Thaler RH. Mental accounting matters. J University Press; 2009.
Behav Decis Mak 1999;12(3):183–206.
Wu G, Zhang J, Gonzalez R. In: Koehler DJ, Har-
22. DellaVigna S, Malmendier U. Paying not to go vey N, editors. Blackwell handbook of judgment
to the gym. Am Econ Rev 2006;96(3):694–719. and decision making. Oxford: Blackwell; 2004.
23. Johnson E, Hershey J, Meszaros J, Kun- pp. 399–423.
reuther H. Framing, probability distortions,
and insurance decisions. J Risk Uncertain
1993;7:35–51.